Circuits and Systems, 2011, 2, 201-209
doi:10.4236/cs.2011.23029 Published Online July 2011 (
Copyright © 2011 SciRes. CS
Chip Design of a Low-Voltage Wideband Continuous-Time
Sigma-Delta Modulator with DWA Technology for WiMAX
Jhin-Fang Huang, Yan-Cheng Lai, Wen-Cheng Lai, Ron-Yi Liu
Department of Electroni c En g i neerin g , National Taiwan University of Science and Technology, Chinese Taipei
Received April 18, 2011; revised May 13, 2011; accepted May 20, 2011
This paper presents the design and experimental results of a continuous-time (CT) sigma-delta (ΣΔ) modula-
tor with data-weighted average (DWA) technology for WiMAX applications. The proposed modulator com-
prises a third-order active RC loop filter, internal quantizer operating at 160 MHz and three DAC circuits. A
multi-bit quantizer is used to increase resolution and multi-bit non-return-to-zero (NRZ) DACs are adopted
to reduce clock jitter sensitivity. The NRZ DAC circuits with quantizer excess loop delay compensation are
set to be half the sampling period of the quantizer for increasing modulator stability. A dynamic element
matching (DEM) technique is applied to multi-bit ΣΔ modulators to improve the nonlinearity of the internal
DAC. This approach translates the harmonic distortion components of a nonideal DAC in the feedback loop
of a ΣΔ modulator to high-frequency components. Capacitor tuning is utilized to overcome loop coefficient
shifts due to process variations. The DWA technique is used for reducing DAC noise due to component
mismatches. The prototype is implemented in TSMC 0.18 um CMOS process. Experimental results show
that the ΣΔ modulator achieves 54-dB dynamic range, 51-dB SNR, and 48-dB SNDR over a 10-MHz signal
bandwidth with an oversampling ratio (OSR) of 8, while dissipating 19.8 mW from a 1.2-V supply. Includ-
ing pads, the chip area is 1.156 mm2.
Keywords: ADC, Analog-to-Digital Conversion, Sigma-Delta Modulator, ΣΔ, DWA
1. Introduction
Sigma-delta modulation techniques have been extended
in moderate and high accuracy analog/mixed-signal IC
applications, such as analog-to-digital data converters
(ADCs), digital-to-analog data converters (DACs), fre-
quency synthesizers, and power amplifiers [1]. Moreover,
ΣΔ modulators are widely used in receivers because of
their ability to provide high-resolution with relatively
low precision components and low power consumption
[2,3]. Oversampling ΣΔ ADCs trade digital signal proc-
essing complexity for relaxed requirements on the analog
components compared to Nyquist-rate ADCs [4]. Due to
the over-sampling characteristics, ΣΔ modulators are
limited on the application of voice band or lower fre-
quency signals. As the ICs process is improved, recently
it makes many researches transfer to wider bandwidth
applications gradually, such as GSM, WCDMA, Blue-
tooth, WiFi, and WiMAX [5]. With the progress of
wireless communication, ADCs need higher OSR in or-
der to achieve higher speed and resolution in the system.
When OSR is programmable, increasing OSR leads to
higher power consumption due to the increased speed
requirement for the integrators and comparators in ΣΔ
modulators. Due to the requirements of low supply volt-
age and low power dissipation in the mobile communica-
tions, the low order ΣΔ modulators of lower SNR are not
suitable for wide bandwidth applications. Therefore, the
high order multi-bit ΣΔ modulator circuit is design to
increase the SNR.
While most of current commercial ΣΔ ADCs for wire-
less applications were implemented by using switched
capacitor (SC) techniques which are also known as dis-
crete-time (DT) ΣΔ ADCs [6-8], mainly due to mature
design methodologies and robustness, more and more
continuous-time (CT) ΣΔ ADCs were reported and
showed impressive performance. Compared with DT
counterparts, the CT ΣΔ ADCs have two main advan-
202 J.-F. HUANG ET AL.
tages. First, the inherent anti-aliasing characteristics of
the CT ΣΔ ADCs reduce the performance requirement of
the anti-aliasing filter further and hence reduce the power
consumption of the transceiver. Second, the bandwidth
requirement of the operational amplifiers (op amps) in
CT ΣΔ ADCs is much lower than that of the op amps in
DT ones for a given sampling rate, so the CT ΣΔ ADCs
are more suitable for broadband applications. Hence we
propose a low-voltage, lower power consumption and
high resolution CT ΣΔ modulator. Our target is to design
a 10 MHz input signal bandwidth and 160MHz sam-
ple-rate ΣΔ modulator implemented in TSMC 0.18 μm
CMOS process.
This paper begins with a brief summary of the innova-
tive CT ΣΔ circuit design. Section 2 introduces the sys-
tem architecture of the wideband CT ΣΔ modulator. Sec-
tion 3 describes the design of building blocks of the
modulator, while Section 4 presents the measured results
of the prototype. Section 5 summarizes the paper.
2. System Circuit Architecture
As high sampling frequency will restrict our design tech-
niques, low OSR is more suitable for the bandwidth of
10 MHz structure. In order to achieve better resolution
and reduce quantization noise, at least a third-order noise-
transfer function is indispensable. Figure 1 shows the
proposed CT ΔΣ modulator architecture which consists
of a 4-bit internal quantizer, operating at 160 MHz with
an OSR of 8, and a third-order single-loop filter. In order
to decrease power consumption and maintain a good alias
filter characteristic, a combination of feedforward and
feedback stabilized loop filters [9] is adopted. The 4-bit
quantizer, including the NRZ feedback DAC is con-
nected to the output of the loop filter. The quantizer de-
lay is set to half of the sampling period. This large delay
is compensated exactly by an additional feedback path
Figure 1. Continuous-time ΣΔ modulator architecture.
A possible design technique for CT modulators is de-
scribed [10,11]. Specifying a DT modulator and trying to
find the equivalent CT modulator between s-plane and
z-plane can use the impulse-invariance transform ex-
pressed as:
 
where RD(s) is the Laplace transform of impulse response
of the DAC and H(z) is the DT loop filter. Equation (1) is
adopted to compensate the impairments of the circuit
such that the resulting CT domain modulator still
matches with the specified DT modulator. The follow-
ings outline the procedure used to determine the direct
feedback coefficients such that quantizer delay is can-
celed exactly. First, a noise-transfer function (NTF) in
the z-domain is chosen, then the loop transfer function
Hloopz(z) is derived as follows:
 
Using the discrete-to-discrete (d2d) function in the
MATLAB control system toolbox can easily transform z
into z1/2 shown in (3):
22 2
LoopZ nn
bzbz b
zaz aza
 0
 
 
After multiplying in the formula, a constant term
can be easily separated from the transfer function
1/ 2
rD z
L of the loop filter as follows:
22 2
filterD nn
bzbz b
zaz aza
 
 
 
Using the discrete-to-continuous (d2c) function in the
MATLAB tool box converts this transfer function
1/ 2
Lz to continuous time. A possible loop filter is
then defined in the CT domain is RC loop filter.
3. Continus-Time ΣΔ Modulator
3.1. Continuous-Time ΣΔ Modulator Circuit
The 4-bit CT ΣΔ modulator circuit including the excess
loop delay compensation is shown in Figure 2 [12]. The
modulator consists of a 4-bit internal quantizer, operating
at 160 MHz with an OSR of 8, and a third-order single-
loop filter. The loop filter is realized as an active RC
filter. Due to the low supply voltage and the high-linear-
ity requirement two-stage op amps with CT common
mode feedback (CMFB) are used. The 4-bit quantizer is
connected to the DWA circuit followed by the feedback
Copyright © 2011 SciRes. CS
Copyright © 2011 SciRes. CS
3.2. Loop Filter
DAC. When multi-bit quantizer is used for better quan-
tization resolution, in-band tones are often observed due
to the element mismatch in the feedback DAC. To solve
the mismatch problem, dynamic element matching is
used in the circuit design. The RC time constant in the
circuit dominating the entire NTF pole function, will
keep stable and therefore the circuit phase margin will
also be stable. Due to the process variation, the capacitor
tuning circuit is used in this modulator. The DAC1 and
DAC2 circuits provide the first and second feedback
paths K1fb and K2fb, respectively. The third K3fb is the
feedback path around the 4-bit quantizer and its output is
connected to the DAC2 output. In order to reduce the
loop filter capacitive loading effects, all the comparators
inside flash ADC input transistors must be the minimum-
There are three types of commonly used CT integrators:
active-RC integrators, Gm-C integrators and MOSFET-C
integrators. In this design, an active-RC integrator is
chosen for the three stages of the third-order loop filter
because it has high linearity and easy interface with
DACs compared to Gm-C integrators. If active-RC inte-
grators were used, resistive loading increases the power
requirements due to the need for buffer stages. A higher
frequency range is additionally demanded in connection
with a high linearity, the active-RC filters are the pre-
ferred structure. The third-order noise shaping loop filter
is realized by an active-RC op amp circuit as shown in
Figure 2. The advantages of this implementation are
high linearity and high output signal swing, and it also
provides a good virtual ground for the modulator feed-
back DACs. This eases design, especially with low sup-
ply voltages. Figure 3 show the architectures of the
1.2-V fully differential op amp [13] and the correspond-
ing CMFB circuit is shown in Figure 4. The op amp
shown in Figure 3 is a two-stage that consists of a folded
cascode input stage, a common source output stage and a
CT CMFB which is similar to a transimpedence ampli-
General “zero-order” feedback path requires additional
summing amplifier and return-to-zero (RZ) DAC con-
tains additional logic control circuits. However, this will
cause additional loop delay, increase power consumption
and complicate the circuit. Therefore, to improve these
drawbacks, in this work, a feedback path is directly con-
nected to the last integrator input, and then the additional
summing amplifier is eliminated. Obviously this way
reduces power consumption and excess loop delay.
Figure 2. Four-bit CT ΣΔ modulator architecture.
Figure 3. Fully differential 1.2-V op amp circuit.
Copyright © 2011 SciRes. CS
As shown in Figure 4, two resistors with values equal
to 2R1 are used to sense the output common mode volt-
age and produce a current I1. This current is compared
with I2, which is set by the desired common-mode volt-
age (VDD/2) and the resistor R1. The difference between I1
and I2 is then converted into a control voltage labeled as
Vcmfb by transistors M3, M6 and M9. The control voltage
will then be used to adjust the VGS’s of M4 and M5, such
that the output common-mode voltage is stabilized to
about VDD/2. The CMFB circuit has advantages of al-
lowing rail-to-rail output swing. Furthermore, it does not
need any level shift or attenuation on the common mode
signal, unlike other CT CMFB circuits that use differen-
tial pairs.
3.3. Four-Bit Flash ADC
In ΣΔ modulators, the main specifications for the quan-
tizer are offset, speed, area and power consumption re-
quirements. Moreover the quantizer has to operate at the
speed required by the oversampling process. Therefore it
must be implemented as a flash ADC [14]. The block
diagram of the 4-bit flash ADC used in the quantizer is
shown in Figure 5. It consists of 15 differential com-
parators, a resistor ladder, and a thermal to binary en-
coder. These comparators compare the input signal with
reference voltages by a resistor ladder biased by the full
scale reference. Consequently, the comparator outputs
constitute a thermometer code, which is converted to
binary by the encoder. Since flash architectures employ
comparators, they are susceptible to metastability errors.
In order to lower the probability of metastable states, the
thermometer-binary decoding can be pipelined so that
potentially indeterminate outputs are allowed more re-
generation time [15].
The clocked comparator is composed of a preamplifier
and a regenerative latch. The schematic is shown in Fig-
ure 6. The comparator utilizes the advantages of the low
kickback noise in static comparators and the high regen-
eration speed in dynamic comparators. On one hand,
keeping the preamplifier continuously biased throughout
the conversion period significantly reduces the kickback
disturbance; on the other hand, the dynamic flip-flop in
the latch circuit will shorten the regeneration and reset
time [16].
Figure 4. Continuous-time CMFB circuit.
Figure 5. Four-bit quantizer and encoder structure.
Figure 6. Schematic of the comparator circuit.
Copyright © 2011 SciRes. CS
3.4. Feedback DACs
As shown in Figure 7, a multi-bit current-steering DAC,
feeds current to the virtual grounds of the active-RC in-
tegrators, therefore good DAC linearity can be achieved.
The improved DAC linearity is another advantage of
active-RC integrators when compared to Gm-C integra-
tors [17].
The CMFB circuit for the DAC along with the differ-
ential pairs connects to the op amp input. This DAC
feedback circuit requires a 0.6-V voltage as a reference
voltage, and this may operate in the virtual ground volt-
age. The reference voltage is connected by an external
power supply. Two current sources inject common-mode
currents to prevent a common-mode offset from appear-
ing at the amplifier virtual grounds. The dynamic per-
formance of current-steering DAC’s is limited by the
feedthrough of the control signals to the output lines. The
coupling of the switching control signals to the output
lines through the parasitic gate-drain capacitance of the
switching transistors is also a source of glitches. The
lower part of Figure 7 is the simplified representation of
the current-steering DAC. In this work, to minimize the
feedthrough to the output lines, the drain of the switching
transistors is isolated from the output lines by adding two
cascaded transistors [18].
3.5. DWA Circuit
Combining ΣΔ modulators with multi-bit quantization is
an effective means to achieve a high dynamic range and
a wide bandwidth. The major obstacle in designing mul-
ti-bit ΣΔ modulators is that good component matching is
required for internal DAC linearity. Good attenuation of
DAC noise due to component mismatches can be pro-
vided by the DWA algorithm, which ideally can achieve
a first-order DAC noise shaping. For DWA to be more
useful in multi-bit SDM’s, the DAC baseband tones must
be removed. Conventionally, the problem is circum-
vented by adding dither. However, adding dither con-
tributes additional noise to the base-band, degrades SNR
and possibly destabilizes the modulator. A low- com-
plexity high-speed circuit is proposed for the implemen-
tation of a DWA technique without adding dither, used
for reducing DAC noise due to component mismatches
The block diagram of the DWA logic is shown in
Figure 8. The input of the DWA logic is connected to
the four-bit quantizer output. The DWA logic converts
the quantizer output code to the control signals, Si, i = 0,
1,, 15, for the element selection of 4-bit DAC. A 4-bit
adder and a 4-bit register produce two indexes which are
converted to two sets of 16-bit thermometer codes by
two binary to thermometer decoders. When the carry
signal of the adder is low, the output control signals are
the mutual XOR of the two 16-bit thermometer codes.
When the carry signal is high, the control signals are the
mutual XNOR of the two 16-bit codes. The DWA algo-
rithm selects DAC components cyclically one by one. No
unit is reselected before all the others are selected.
3.6. Time Constant Tuning Circuit
CMOS technologies usually do not have tight control
over absolute values of R and C, so an automatic RC
time constant tuning circuit is needed to ensure the ΣΔ
modulator stability and SNR performance over large RC
time constant variations. Therefore, a discrete capacitor
tuning scheme is employed to calibrate the time constant
of the active-RC integrators. The adjustable capacitor
array is shown in Figure 9.
Figure 7. A multi-bit current-steering DAC schematic.
206 J.-F. HUANG ET AL.
Figure 8. The block diagram for the DWA realization.
Figure 9. Tunable capacitor array.
The capacitors in the arrays are binary-weighted ex-
cept the “always-in-use” capacitor which is equal to the
most significant bit (MSB) capacitor, 8C. This sizing
method provides constant tuning steps with the least
number of capacitors. The 3-bit digital control codes are
fed externally to choose which capacitors to use.
4. Measurement Results
The proposed third-order multi-bit CT ΣΔ modulator in
this paper is implemented in TSMC 0.18-μm CMOS
process. Post-processing was performed using MATLAB
before tapout. The modulator samples signals at 160 MHz
with 10 MHz signal bandwidth and oversampling ratio of
8 and operates with a 1.2 V supply voltage. The total
power consumption is 19.8 mW. Figure 10 shows the
modulator die microphotograph including the wire
bounding pads. The CT ΣΔ modulator is essentially a
mixed-signal system which includes integrator, quantizer,
and digital circuits. To achieve high resolution and line-
arity, caution should be taken in the layout design to re-
duce the effects of mismatch, parasitic and digital noise
coupling to analog blocks. The total chip area including
bonding pads is 0.9 × 1.284 mm2.
The DWA circuit is located on the left of the chip. The
noisy clock generator is placed in the bottom right-hand
0.9 mm
1.284 m
Figure 10. Microphotograph of the CT ΣΔ modulator.
nalog blocks. A single-to-differential circuit converts a
the single-ended input signal to a balanced differential
signal input to the ADC. The output data stream of the
modulator was captured using a logic analyzer. Figure
11 shows the digital outputs of the CT ΣΔ modulator
measured by the logic analyzer for an input sinusoid at 3
MHz. The output spectrum density of the CT ΣΔ modu-
lator analyzed by logic analyzer for an input sinusoidal
signal of 3 MHz is shown in Figure 12. A peak SNDR of
48 dB which corresponds to a 7.7-bit within a bandwidth
of 10 MHz is measured. The measured SNR and SNDR
versus input signal level of the CT ΣΔ modulator for an
input sinusoid at 3 MHz are plotted in Figure 13. The
measured input peak dynamic range is 54 dB. Figure 14
summarizes the measured SNR and SNDR for varying
input frequencies. The SNR and SNDR fall to 46 dB and
43 dB respectively for a 9 MHz input signal. Because the
integrator is basically a low pass filter, the modulator
acts as low-pass filtering characteristic. When input fre-
quency is increased, the SNR/SNDR values will be de-
side to prevent interference with the weakly sensitive
Copyright © 2011 SciRes. CS
Figure 11. Measured digital output of the CT ΣΔ modulator at f = 3 MHz.
Figure 12. Measured output spectrum density.
Figure 13. Measured SNR and SNDR vs. input signal level.
Copyright © 2011 SciRes. CS
208 J.-F. HUANG ET AL.
Figure 14. Measured SNR and SNDR vs. input frequency.
The performance parameters of this chip are summa-
Table 2. Performance comparisons with other literatures.
ed in Table 1. In this work, we use the design strategy
for low-power CT ΣΔ modulator proposed by [20]. This
concept is based on the figure of merit (FOM) which
takes the overall power consumption, the dynamic range,
and the signal bandwidth into account to find the most
power-efficient ΣΔ modulator implementation with re-
spect to these design parameters.
The FOM used is defined as
where P(mW) represents the power consumption, B is
able 1. The performance summary of the CT ΣΔ modulator.
Parameters Measured results
number of bits and fB(MHz) is the bandwidth. The per-
formance comparisons with other literatures are shown in
Table 2. The smaller the FOM value is, the better the
overall performance is. From this comparison table, it is
confirmed that the proposed modulator with low voltage
operations can achieve a wide bandwidth and lower
power consumption.
Sampling Frequency 160 MHz
Signal Bandwidth 10 MHz
SNR 51 dB
Parameter This work[21] [22] [23] [24]
Te0. 0. 0.chnology 0.18 um18 um18 um90 um13 um
(V) 1.2 1.8 1.2 1.2 2.5
BW (MHz) 10 10 7.5 1.92 100
SNR (dB) 51 63 71 66.4 58.9
SNDR (dB) 48 56 67 62.4 53.1
ENOB (Bit7.7 9 10.8 10.1 8.5
Power (mW) 19.8 22.4 89 12.5 350
FOM (pJ/Conv4.76 1.953.33 2.97 4.83
c Range
ion 1
ENOB 7.7 bits
Dynami54 dB
Power Supply 1.2 V
Power Dissipat9.8 mW
Chip Area 1.156 mm2
Process C 0.18 um CM
-power and wide bandwidth CT ΣΔ
odulator has been implemented in a TSMC 0.18-um
ank the staff of the CIC for
. Conclusio
A low-voltage, low
technology. The low-complexity high-speed implemen-
tation of the DWA technique for the reduction of base-
band tones is used in this modulator. The excess loop
delay set to half the sampling period of the quantizer has
been used to avoid degradation of modulator stability in
this architecture. All integration capacitors are tunable to
overcome time constant variation. In addition, CT ΣΔ
modulator provides a significant amount of inherent an-
ti-aliasing, which is especially important when OSR is
minimized in order to maximize the input bandwidth.
The CT ΣΔ modulator itself occupies just 1.16 mm2 and
consumes 19.8 mW. The modulator achieves a SNR of
51 dB and SNDR of 48 dB over 10 MHz signal band-
6. Ack
The authors would like to th
Copyright © 2011 SciRes. CS
e chip fabrication and technical supports with
. Zhao, “Continuous-Time Sigma-Delta
Modulator Design for Low Power Communication
oermund, “A 3.3-mW ΣΔ Modulator for
th the
number of T18-98D-158.
7. References
[1] J. Yu and B
plications,” Proceedings of IEEE ASICON, October 2007,
pp. 715-720.
[2] R. H. M. van Veldhoven, B. J. Minnis, H. A. Hegt and A.
H. M. van R
UMTS in 0.18-um CMOS with 70-dB Dynamic Range in
2-MHz Bandwidth,” IEEE Journal of Solid-State Circuits,
Vol. 37, No. 12, 2002, pp. 1645-1652.
[3] L. Dorrer, F. Kuttner, P. Greco, P. Tor
“A 3-mW 74-dB SNR 2-MHz
ta and T. Hartig
Continuous-Time D
sigma ADC with a Tracking ADC Quantizer in 0.13-um
CMOS,” IEEE Journal of Solid-State Circuits, Vol. 40,
No. 12, 2005, pp. 2416-2427.
[4] S. R. Norsworthy, R. Schreier
ta-Sigma Data Converters: Theo
and G. C. Temes, “Del-
ry, Design and Simula-
eless Applications,”
a Pipeline
tion,” IEEE Press, New York, 1996.
[5] B. J. Farahani and M. Ismail, “Adaptive Sigma Delta
ADC for WiMAX Fixed Point Wir
IEEE Midwest Symposium on Circuits and Systems, Vol.
1, Covington, 7-10 August 2005, pp. 692-695.
[6] T. L. Brooks, D. H. Robertson, D. F. Kelly, A. D. Muro
and S. W. Harston, “A Cascaded Sigma-Delt
A/D Converter with 1.25 MHz Signal Bandwidth and 89
dB SNR,” IEEE Journal of Solid-State Circuits, Vol. 32,
No. 12, 1997, pp. 1896-1906. doi:10.1109/4.643648
[7] R. Jiang and T. Fiez, “A 14-bit Delta-Sigma ADC with
8X OSR and 4-MHz Conversion Bandwidth in A 0.18 um
CMOS Process,” IEEE Journal of Solid-State Circuits,
Vol. 39, No. 1, 2004, pp. 63-74.
[8] A. Bosi, A. Panigada, G. Cesura
80 MHz 4x Oversampled Cascaded
and R. Castello, “An
ΔΣ-Pipelined ADC
ators for High-Speed A/D Conver-
ara, “A 70-mW 300
with 75 dB DR and 87 dB SFDR,” IEEE International
Solid-State Circuits Conference, 2005, pp. 174-175.
[9] F. Munoz, K. Philips and A. Torralba, “A 4.7 mW 89.5 dB
DR CT Complex DS ADC with Built-In LPF,” IEEE
ternational Solid-State Circuits Conference, February
2005, pp. 500-501.
[10] J. A. Cherry and W. M. Snelgrove, “Continuous-Time
Delta-Sigma Modul
sion,” Kluwer, Boston, 2000.
[11] S. Paton, A. D. Giandomenico, L. Hernandez, A. Wies-
bauer, T. Potscher and M. Cl-MHz
CMOS Continuous-Time ΣΔ ADC with 15-MHz Band-
width and 11 Bits of Resolution,” IEEE Journal of Sol-
id-State Circuits, Vol. 39, No. 7, July 2004, pp. 1056-
1063. doi:10.1109/JSSC.2004.829925
[12] G. Mitteregger, C. Ebner. S. Mechnig, T. Blon, C. Holu-
igue and E. Romani, “A 20-mW 640-MHz C
tinuous-Time ΣΔ ADC with 20-MHz Signal Band- width,
80-dB Dynamic Range and 12-Bit ENOB,” IEEE Journal
of Solid-State Circuits, Vol. 41, No. 12, 2006, pp.
2641-2649. doi:10.1109/JSSC.2006.884332
[13] S. Karthikeyan, S. Mortezapour, A. Tammineedi and E.
Lee, “Low-Voltage Analog Circuit Design Based on Bi-
ased Inverting Opamp Configuration,” IEEE Transac-
tions on Circuits and Systems II: Analog and Digital
Signal Processing, Vol. 47, No. 3, 2000, pp. 176-184.
[14] L. Lamarre, M. Louerat and A. Kaiser, “A Simple 3.8 mW
300 MHz, 4-Bit Flash A
nalog-to-Digital Converter,” Pro-
ceedings of the SPIE, Vol. 5837, No. 51, 2005, pp. 825-
832. doi:10.1117/12.608343
[15] B. Razavi, “Principles of Data Conversion System De-
sign,” IEEE Press, New York, 1995.
ed Comparator for
[16] J. Chen, S. Kurachi, S. Shen, H. Liu, T. Yoshimasu and Y.
Suh, “A Low-Kickback-Noise Latch
High-Speed Flash Analog-to-Digital Converters,” Inter-
national Symposium on Communications and Information
Technologies, Beijing, 12-14 October 2005, pp. 250-253.
[17] Z. Li and T. S. Fiez, “A 14-Bit Continuous-Time Delta-
sigma A/D Modulator with 2.5 MHz Signal Bandwidth,”
IEEE Journal of Solid-State Circuits, Vol. 42, No. 9,
2007, pp. 1873-1883. doi:10.1109/JSSC.2007.903086
[18] J. Bastos, A. M. Marques, M. S. J. Steyaert and W. San-
sen, “A 12-Bit Intrinsic Accuracy High-Speed CMOS
DAC,” IEEE Journal of Solid-State Circuits, Vol. 33, No.
12, 1998, pp. 1959-1969. doi:10.1109/4.735536
[19] T. H. Kuo, K. D. Chen and H. R. Yeng, “A Wideband
CMOS Sigma-Delta Modulator with Incremental Data
Weighted Averaging,” IEEE Journal of Solid-State Cir-
cuits, Vol. 37, No. 1, 2002, pp. 11-17.
[20] F. Gerfers, M. Ortmanns and Y. Manoli
Power-Efficient Contin
, “A 1.5-V 12-Bit
uous-Time Third-Order ΣΔ Mod-
ulator,” IEEE Journal of Solid-State Circuits, Vol. 38, No.
8, 2003, pp. 1343-1352. doi:10.1109/JSSC.2003.814432
[21] L. J. Breems, R. Rutten and G. Wetzker, “A Cascaded
Continuous-Time ΣΔ Modulator with 67-dB Dynamic
Range in 10-MHz Bandwidth,” IEEE Journal of Sol-
id-State Circuits, Vol. 39, No. 12, 2004, pp. 2152-2160.
[22] S. D. Kulchycki, R. Trofin, K. Vleugels and B. A. Woo-
ley, “A 77-dB Dynamic Range, 7.5-MHz Hybrid Con-
tinuous-Time/Discrete-Time Cascaded ΣΔ Modulator,”
IEEE Journal of Solid-State Circuits, Vol. 43, No. 4,
2008, pp. 796-804. doi:10.1109/JSSC.2008.917499
[23] M. Anderson and L. Sundstrom, “Design and Measure-
ment of a CT ΔΣ ADC with Switched-Capacitor Switched-
Resistor Feedback,” IEEE Journal of Solid-State Circuits,
Vol. 44, No. 2, 2009, pp. 473-483.
[24] A. Hart and S. P. Voinigescu, “A 1 GHz Bandwidth Low-
Pass ΔΣ ADC with 20-50 GHz
Adjustable Sampling
Rate,” IEEE Journal of Solid- State Circuits, Vol. 44, No.
5, 2009, pp. 1401-1414. doi:10.1109/JSSC.2009.2015852
MOS Con-
Copyright © 2011 SciRes. CS